PyWiFeS

modified by jhwshin

Python Packages and PATHS are contained within the python virtual environment pipeline for modularity.

---

Requirements:

• python3
• pip

---

ENV Setup:

0. Create a python virtual environment:

$   python -m venv venv

1. Patch activate to add pipeline/{src,reduction_scripts} to PYTHONPATHS in venv:

$   patch -u venv/bin/activate -i pypaths.patch

2. Create shortcut to activate venv:

$   ln -sv venv/bin/activate .

3. Activate to start the python venv:

$   source activate

4. Upgrade pip to latest version:

$   pip install --upgrade pip

5. Install required packages with pip:

$   pip install -r requirements.txt

---

To EXIT the venv simply type:

$   deactivate

To START the environment again type:

$   source activate

---

Pipeline Instructions:

0. All files are contained within template folder, copy it and rename to any workspace name:

$   cp template <NAME>
$   cp <PATH_TO_FITS>/*.fits <NAME>/raw_data

1. Generate metadata from raw fits files:

$   python2 generate_metadata_script.py raw_data

• raw_data dir containing *.fits files

Outputs: save_{blue,red}_metadata.py files

2. Create pickle files:

$   python2 save_{red,blue}_metadata.py

Outputs: wifes{B,R}_.*_metadata.pkl files

3. Run reduce_{blue,red}

$   python2 reduce_{blue,red} wifes{B,R}_.*_metadata.pkl

• Or run this command for sanitised output with log file:

$   python2 -u -W ignore reduce_{blue,red} wifes{B,R}_.*_metadata.pkl 2>&1 | tee log.txt

# USEFUL COMMANDS:

# -u              unbuffered prints immediately (otherwise will print sporadically with tee)
# -W ignore       ignore Warnings
# 2>&1            prints stderr to stdout (used for tee)

---

Changes / Fixes:

Fixes:

• new reference_data from latest origin PyWiFeS in github
• rename all reference_data files to .tab as well as stdstar_lookup_table for consistency
• fixed some bugs in src/wifi_calib.py when searching for star ref_names
• renamed template folder and added default reduc_{b,r} folder

Features:

• now running pipeline in pyvenv for better abstraction with contained packages and pythonpaths
• added sort.py as a manual inspection tool
• added this README